Project Summary More than 26 million Americans have diabetes, which is characterized by progressive vascular dysfunction accompanied by destabilization of the microvasculature. Tissues exposed to diabetic conditions exhibit reduced vascular density and reduced pericyte (PC) density, a key vascular effector cell that wraps around blood vessels, regulating and providing stability to the microvasculature. While a reduction of PCs (often described as ?PC dropout?) has been theorized to be a significant contributor to degradation of the microvasculature in diabetes, little is known of this process or how it initiates. PC dropout has been studied in the retina, considered a ?canary in the coal mine? for revealing diabetes-induced vascular pathology. The retina is advantageous for studying PCs because of its multi-layered, planar microvasculature and ease of imaging cell-level details. Studying PC behaviors during diabetes-induced vascular destabilization in the retina will give insight into the progression and future treatments of diabetic microvascular pathologies. We have recently identified a subset of PCs in the retina that exhibit atypical morphology, such as off-vessel cell bodies and cell processes extending to neighboring capillaries, and we have observed that these atypical PCs are significantly more abundant in diabetic conditions. We hypothesize that this atypical morphology marks PCs that are in the early stages of dropping out and represents an intermediate PC subpopulation that can be therapeutically targeted to prevent diabetes-induced vascular dropout. We will test this hypothesis by combining agent-based computational modeling of the mouse retina with in vivo imaging of PCs in the limbal vessels of the cornea. In Aim 1, we will conduct a time course study of pericyte dynamics following pharmacological stimulation of the PDGF, Notch, and Ang2 pathways, acquire time lapse video of PC behaviors, and develop a retinal agent based model (ABM) that simulates this process. In Aim 2, we will use the ABM to predict a therapeutic regimen that reduces the proportion of PCs with atypical morphology in diabetic mice. This research will produce new insights into how the microvasculature adapts in the early stages of diabetes and test methods to stabilize vasculature by targeting PCs. Pericytes are found in every vascularized tissue in the body, and their dropout could contribute to the deficient wound healing ability and other microvascular complications that diabetics experience. In uniting experimental studies with computational modeling, this project will yield novel molecular mechanisms that could be targeted to prevent or reverse a key initiating step in microvascular maladaptation during diabetes. !